xlmr-large-all-CLS-ET / test_eval_ar.txt
HHansi's picture
Upload folder using huggingface_hub
80432ee verified
Default classification report:
precision recall f1-score support
F 0.8511 0.7200 0.7801 500
T 0.7574 0.8740 0.8115 500
accuracy 0.7970 1000
macro avg 0.8042 0.7970 0.7958 1000
weighted avg 0.8042 0.7970 0.7958 1000
ADJ
Accuracy = 0.7346938775510204
Weighted Recall = 0.7346938775510204
Weighted Precision = 0.7688370873382583
Weighted F1 = 0.7311342146921662
Macro Recall = 0.7463312368972745
Macro Precision = 0.7603012848914488
Macro F1 = 0.7329140461215934
ADV
Accuracy = 0.8
Weighted Recall = 0.8
Weighted Precision = 0.64
Weighted F1 = 0.7111111111111111
Macro Recall = 0.5
Macro Precision = 0.4
Macro F1 = 0.4444444444444445
NOUN
Accuracy = 0.8076923076923077
Weighted Recall = 0.8076923076923077
Weighted Precision = 0.8111648454026467
Weighted F1 = 0.8069870256602342
Macro Recall = 0.8069999999999999
Macro Precision = 0.8115506829260177
Macro F1 = 0.8068302963197735
VERB
Accuracy = 0.7989949748743719
Weighted Recall = 0.7989949748743719
Weighted Precision = 0.8072385648560104
Weighted F1 = 0.7978523501915108
Macro Recall = 0.799782811829179
Macro Precision = 0.8067233404821581
Macro F1 = 0.7979951782768684